The overall goals of this proposal are to develop and validate a novel magnetic resonance imaging method for mapping tissue protein and creatine content in order to detect and characterize tumor cellular and metabolic features. The ability to assess such tumor attributes will have a broad range of both clinical and preclinical applications, particularly since such fundamental biophysical characteristics are highly dissimilar to that found in the surrounding normal tissue and could serve as improved early indicators of treatment response as compared to conventional assays of gross tumor volume, which often change much later in the course of therapy. The proposed methods for achieving this objective include a significant and novel reformulation of chemical exchange saturation transfer (CEST) imaging. There are inherent and critical technical obstacles to the widespread application of CEST methods and major limitations on their use in practice. CEST suffers from a lack of quantification and is highly susceptible to artifacts originating from static field inhomogeneities lipid content, spectral overlap of multiple metabolites signals, and the inherently asymmetric background macromolecular resonance. Chemical exchange rotation transfer (CERT) is our proposed reformulation of CEST that overcomes the shortcomings of CEST. Furthermore, CERT can act as an exchange rate filter and can separate the contribution of creatine from that of other overlapping metabolites. We will 1) develop, optimize, and validate CERT imaging methods; 2) quantify tissue changes in protein and creatine content in rat models of tumor growth and remission; and 3) quantify tissue changes in protein and creatine content in cancer patients treated with Bevacizumab, which restores the blood brain barrier and eliminates inflammation, and hence makes conventional post-contrast T1 and diffusion imaging uninformative.